Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A system for crack detection, the system comprising: a storage device; and a processing system connected to the storage device; and a program stored in the storage device, wherein execution of the program by the processing system causes the system to perform functions, including functions that: (i) establish an appropriate structuring element based on a working distance and focal length of a three-dimensional (3D) structure of a scene; from a plurality of images of the scene; (ii) segment potential crack patterns by applying a morphological operation; (iii) determine appropriate features for each segmented pattern; and (iv) classify a crack from a non-crack pattern using a trained classifier, thereby forming a multiscale crack map, wherein: for the plurality of images of the scene there are a plurality of data acquisition parameters, wherein the relation between different image acquisition parameters is in accordance with the following: SF = ( WD FL ) ( SS SR ) n , where feature size (SF) is the size of a crack thickness represented by n pixels in an image, working distance (WD) is the working distance between camera and object, focal length (FL) is the camera focal length, sensor size (SS) (mm) is the camera sensor size, and sensor resolution (SR) is the camera sensor resolution in pixels; or the multiscale crack map is formulated in accordance with the following: and 0 otherwise; where J m is the crack map at scale m of a structuring element, S min , is the minimum structuring element size, C k is the binary crack image obtained by using k as the structuring element, and u and v are the pixel coordinates of the crack map image.
A crack detection system uses a computer with a storage device and a processor. A program runs on the processor to detect cracks in images of a 3D scene. The program first determines the appropriate size and shape of a "structuring element" (a small image used for morphological operations) based on the camera's distance to the object and its focal length. It then uses this structuring element to find potential crack patterns in the images through a morphological operation (image processing that alters the structure of objects). Next, it determines key features for each potential crack. Finally, a pre-trained classifier (like a machine learning model) distinguishes real cracks from other patterns, creating a multi-scale crack map. The system accounts for different image acquisition parameters using the formula: SF = (WD/FL) * (SS/SR) * n, where SF is feature size (crack thickness in pixels), WD is working distance, FL is focal length, SS is sensor size, SR is sensor resolution, and n is the number of pixels representing the crack. The multi-scale crack map combines crack images obtained using different structuring element sizes.
2. The system of claim 1 , wherein the trained classifier comprises a neural network.
The crack detection system described previously uses a neural network as the pre-trained classifier to distinguish real cracks from other patterns. The system determines the appropriate size and shape of a "structuring element" based on the camera's distance to the object and its focal length. It then uses this structuring element to find potential crack patterns in the images through a morphological operation. Next, it determines key features for each potential crack. The neural network is trained to classify cracks, thereby forming a multi-scale crack map. The system accounts for different image acquisition parameters using the formula: SF = (WD/FL) * (SS/SR) * n, where SF is feature size, WD is working distance, FL is focal length, SS is sensor size, SR is sensor resolution, and n is the number of pixels representing the crack. The multi-scale crack map combines crack images obtained using different structuring element sizes.
3. The system of claim 1 , wherein the trained classifier comprises a support vector machine (SVM).
The crack detection system described previously uses a support vector machine (SVM) as the pre-trained classifier to distinguish real cracks from other patterns. The system determines the appropriate size and shape of a "structuring element" based on the camera's distance to the object and its focal length. It then uses this structuring element to find potential crack patterns in the images through a morphological operation. Next, it determines key features for each potential crack. The SVM is trained to classify cracks, thereby forming a multi-scale crack map. The system accounts for different image acquisition parameters using the formula: SF = (WD/FL) * (SS/SR) * n, where SF is feature size, WD is working distance, FL is focal length, SS is sensor size, SR is sensor resolution, and n is the number of pixels representing the crack. The multi-scale crack map combines crack images obtained using different structuring element sizes.
4. The system of claim 1 , wherein the trained classifier comprises a neural nearest-neighbor classifier.
The crack detection system described previously uses a neural nearest-neighbor classifier as the pre-trained classifier to distinguish real cracks from other patterns. The system determines the appropriate size and shape of a "structuring element" based on the camera's distance to the object and its focal length. It then uses this structuring element to find potential crack patterns in the images through a morphological operation. Next, it determines key features for each potential crack. The nearest-neighbor classifier is trained to classify cracks, thereby forming a multi-scale crack map. The system accounts for different image acquisition parameters using the formula: SF = (WD/FL) * (SS/SR) * n, where SF is feature size, WD is working distance, FL is focal length, SS is sensor size, SR is sensor resolution, and n is the number of pixels representing the crack. The multi-scale crack map combines crack images obtained using different structuring element sizes.
6. The system of claim 5 , wherein execution of the program by the processor further causes the system to perform a function that: eliminates extracted patterns if the length of a segmented pattern is less than a minimum length specified by a user.
Building on a crack detection system that uses a computer to find cracks in images of a 3D scene by establishing a structuring element, segmenting potential crack patterns, determining features, and using a trained classifier to form a multiscale crack map; the system is further enhanced to eliminate extracted patterns if the length of a segmented pattern is less than a minimum length specified by the user. This filtering step removes noise and irrelevant patterns based on a user-defined threshold for crack length.
7. The system of claim 5 , wherein execution of the program by the processor further causes the system to perform a function that: converts minimum length of interest in unit length to minimum length in pixels in accordance with the following: l p = ( FL WD ) l , wherein l is the defined length by the user in unit length, focal length (FL) is in pixels and working distance (WD) is in unit length, and l p is the length in pixels.
Building on a crack detection system that uses a computer to find cracks in images of a 3D scene by establishing a structuring element, segmenting potential crack patterns, determining features, and using a trained classifier to form a multiscale crack map; the system is further enhanced to convert a user-defined minimum length of interest from real-world units (e.g., millimeters) to pixels. It uses the formula: lp = (FL/WD) * l, where l is the user-defined length in real-world units, FL is the focal length in pixels, WD is the working distance in real-world units, and lp is the equivalent length in pixels. This ensures the user-defined minimum length is accurately applied in the image processing steps.
8. The system of claim 1 , wherein for the plurality of images of the scene there are a plurality of data acquisition parameters, wherein the relation between different image acquisition parameters is in accordance with the following: SF = ( WD FL ) ( SS SR ) n , where feature size (SF) is the size of a crack thickness represented by n pixels in an image, working distance (WD) is the working distance between camera and object, focal length (FL) is the camera focal length, sensor size (SS) (mm) is the camera sensor size, and sensor resolution (SR) is the camera sensor resolution in pixels.
A crack detection system considers the relationship between image acquisition parameters. The system uses the formula: SF = (WD/FL) * (SS/SR) * n, where SF is the feature size (the size of a crack thickness represented by n pixels in an image), WD is the working distance between the camera and the object, FL is the camera focal length, SS is the camera sensor size in millimeters, and SR is the camera sensor resolution in pixels. This formula relates how the size of a crack in the real world translates to its size in the image, based on camera settings and sensor properties.
9. The system of claim 8 , wherein SF = ( WD FL ) n , wherein FL is in pixels.
The crack detection system described previously simplifies the relationship between image acquisition parameters to: SF = (WD/FL) * n, where SF is the feature size (the size of a crack thickness represented by n pixels in an image), WD is the working distance between the camera and the object, FL is the camera focal length (in pixels), and n is the number of pixels representing the crack. This simplified formula is used to relate the real-world crack size to its size in the image based on the camera's focal length and distance.
12. The method claim 11 , wherein the trained classifier comprises a neural network.
In a method of crack detection that establishes a structuring element based on working distance and focal length, segments potential crack patterns using a morphological operation, determines features for each segmented pattern, and classifies cracks using a trained classifier to form a multiscale crack map; the trained classifier comprises a neural network.
13. The method of claim 11 , wherein the trained classifier comprises a support vector machine (SVM).
In a method of crack detection that establishes a structuring element based on working distance and focal length, segments potential crack patterns using a morphological operation, determines features for each segmented pattern, and classifies cracks using a trained classifier to form a multiscale crack map; the trained classifier comprises a support vector machine (SVM).
14. The method of claim 11 , wherein the trained classifier comprises a neural nearest-neighbor classifier.
In a method of crack detection that establishes a structuring element based on working distance and focal length, segments potential crack patterns using a morphological operation, determines features for each segmented pattern, and classifies cracks using a trained classifier to form a multiscale crack map; the trained classifier comprises a neural nearest-neighbor classifier.
16. The method of claim 15 , wherein execution of the program by the processor further configures the system to perform a function to: eliminate extracted patterns if the length of a segmented pattern is less than a minimum length specified by a user.
In a method of crack detection that establishes a structuring element based on working distance and focal length, segments potential crack patterns using a morphological operation, determines features for each segmented pattern, and classifies cracks using a trained classifier to form a multiscale crack map; the method further eliminates extracted patterns if the length of a segmented pattern is less than a minimum length specified by a user.
17. The method of claim 15 , wherein execution of the program by the processor further configures the system to perform a function to: convert minimum length of interest in unit length to minimum length in pixels in accordance with the following: l p = ( FL WD ) l , wherein l is the defined length by the user in unit length, focal length (FL) is in pixels and working distance (WD) is in unit length, and l p is the length in pixels.
In a method of crack detection that establishes a structuring element based on working distance and focal length, segments potential crack patterns using a morphological operation, determines features for each segmented pattern, and classifies cracks using a trained classifier to form a multiscale crack map; the method further converts minimum length of interest in unit length to minimum length in pixels in accordance with the following: l p = ( FL WD ) l , wherein l is the defined length by the user in unit length, focal length (FL) is in pixels and working distance (WD) is in unit length, and l p is the length in pixels.
18. The method of claim 11 , wherein the relation between different image acquisition parameters is in accordance with the following: SF = ( WD FL ) ( SS SR ) n , where feature size (SF) is the size of a crack thickness represented by n pixels in an image, working distance (WD) is the working distance between camera and object, focal length (FL) is the camera focal length, sensor size (SS) (mm) is the camera sensor size, and sensor resolution (SR) is the camera sensor resolution in pixels.
In a method of crack detection that establishes a structuring element based on working distance and focal length, segments potential crack patterns using a morphological operation, determines features for each segmented pattern, and classifies cracks using a trained classifier to form a multiscale crack map; the relation between different image acquisition parameters is in accordance with the following: SF = ( WD FL ) ( SS SR ) n , where feature size (SF) is the size of a crack thickness represented by n pixels in an image, working distance (WD) is the working distance between camera and object, focal length (FL) is the camera focal length, sensor size (SS) (mm) is the camera sensor size, and sensor resolution (SR) is the camera sensor resolution in pixels.
19. The method of claim 18 , wherein SF = ( WD FL ) n , wherein FL is in pixels.
In a method of crack detection that establishes a structuring element based on working distance and focal length, segments potential crack patterns using a morphological operation, determines features for each segmented pattern, and classifies cracks using a trained classifier to form a multiscale crack map; the relation between different image acquisition parameters is in accordance with the following: SF = ( WD FL ) n , wherein FL is in pixels.
Unknown
October 28, 2014
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.